From 0f9024d49f7622d1c135aa2e3f9ddc6bc4349fb9 Mon Sep 17 00:00:00 2001
From: Patrick Simianer <p@simianer.de>
Date: Sat, 3 Dec 2011 21:38:59 +0100
Subject: new rule count

---
 dtrain/hstreaming/rule_count/rulecount.rb | 11 +++++++++++
 1 file changed, 11 insertions(+)
 create mode 100644 dtrain/hstreaming/rule_count/rulecount.rb

(limited to 'dtrain/hstreaming/rule_count/rulecount.rb')

diff --git a/dtrain/hstreaming/rule_count/rulecount.rb b/dtrain/hstreaming/rule_count/rulecount.rb
new file mode 100644
index 00000000..035bdf06
--- /dev/null
+++ b/dtrain/hstreaming/rule_count/rulecount.rb
@@ -0,0 +1,11 @@
+STDIN.set_encoding 'utf-8'
+STDOUT.set_encoding 'utf-8'
+
+while line = STDIN.gets
+  a = line.strip.chomp.split "\t"
+  a[3..a.size].each { |r|
+    id = r.split("|||")[0..2].join("|||").to_s.strip.gsub("\s", "_")
+    puts "#{id}\t1"
+  }
+end
+
-- 
cgit v1.2.3